add var to control only print warning once when blocked

[jackalcooper]

No description provided.

[github-actions]

CI failed, removing label automerge

[github-actions]

Speed stats:

GPU Name: GeForce GTX 1080 

PyTorch resnet50 time: 140.6ms (= 7031.9ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 128.4ms (= 6422.2ms / 50, input_shape=[16, 3, 224, 224], backward is enabled)
Relative speed: 1.09 (= 140.6ms / 128.4ms)

PyTorch resnet50 time: 85.4ms (= 4268.7ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 74.8ms (= 3738.0ms / 50, input_shape=[8, 3, 224, 224], backward is enabled)
Relative speed: 1.14 (= 85.4ms / 74.8ms)

PyTorch resnet50 time: 59.8ms (= 2991.4ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 47.6ms (= 2381.0ms / 50, input_shape=[4, 3, 224, 224], backward is enabled)
Relative speed: 1.26 (= 59.8ms / 47.6ms)

PyTorch resnet50 time: 48.4ms (= 2418.2ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 39.1ms (= 1952.6ms / 50, input_shape=[2, 3, 224, 224], backward is enabled)
Relative speed: 1.24 (= 48.4ms / 39.1ms)

PyTorch resnet50 time: 43.9ms (= 2197.5ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
OneFlow resnet50 time: 37.3ms (= 1866.0ms / 50, input_shape=[1, 3, 224, 224], backward is enabled)
Relative speed: 1.18 (= 43.9ms / 37.3ms)

[chengtbf]

这个有测试过么，我们这里实验的结果是好像还会多次发

[jackalcooper]

这个有测试过么，我们这里实验的结果是好像还会多次发

有的，还是多次应该是不同iter，如果想完全去掉就设置

 export GLOG_minloglevel=3

[chengtbf]

这个间隔能否延长到 10s ？

[jackalcooper]

这个间隔能否延长到 10s ？

刚才和新奇讨论过了，10s太长了，会掩盖问题。

[jackalcooper]

这个间隔能否延长到 10s ？

现在是5s已经相当长了

[chengtbf]

好的